How we report

The rules behind the journalism, the rubric behind the rankings.

Sourcing

Every load-bearing claim carries a source and a date. We draw on the sector's primary research — the MemberWise Digital Excellence Report, ASAE's State of Associations, ASI's Membership Performance Benchmark — plus vendor documentation, published pricing and our own reporting. Pricing sentences carry an "as of" month, because pricing changes and pages don't always keep up.

The ranking rubric

Ranked lists score every entry on the same five axes. Default weights: capability 25%, fit 25%, governance 20%, cost transparency 15%, implementation 15%. A list may reweight an axis for its readership (our UK professional-bodies ranking double-weights fit); when it does, the weights used are printed in that list's scorecard, alongside per-axis grades and the weighted totals, so the ordering can be reproduced and challenged. Where a ranking compares combinations rather than products (our AI piece ranks five stacks, each an engine plus a layer plus education), the same axes are applied to the combination as a whole and the scorecard says so.

• Capability — what the product demonstrably does today, not what the roadmap promises. Vendor-published claims count, and are labelled as vendor-published.
• Fit for membership organisations — evidence of use by associations, professional bodies or unions, not generic mid-market CRM claims.
• Governance & safety — permissions, auditability, human approval where the tool acts on live data.
• Cost transparency — published pricing scores above quote-only opacity.
• Implementation reality — the effort between contract and value, as reported by the market.

Every entry gets real cons and a "not for you if" line. Products that decline to publish pricing are told so in print. No vendor pays to appear, move up, or stay off a list, and rankings are revisited when readers challenge the working: our AI ranking was restructured in August 2026 after exactly such a challenge, from a ten-product list into a market map and five stacks.

Dates and updates

Published and updated dates are visible on every piece and match the page's structured data. We bump an updated date only for substantive changes — a re-ranking, corrected figures, new sections — so the dateline is the record that matters.

Corrections

We correct in place, note the correction at the foot of the piece, and update the dateline. Spot an error: tell us.

AI crawlers

Our content is open to search engines and AI answer engines alike; we'd rather be cited accurately than scraped badly. The full policy lives in our robots.txt.